User-agent: Mozilla/5.0 (Windows NT 10.0; WOW64; rv:68.0) Gecko/20100101 Thunderbird/68.12.0
I wonder, could PMC consider securing their consensus as a
prerequisite to it's vote? Might avoid having multiple places for
approval and the possibility of receiving differing results.
-- Ed
On 9/21/2020 11:40 AM, EMO EMO wrote:
Ivar wears two hats.
As a representative of the PMC, Ivar votes as a member of
the specification committee on their ballot on behalf of the
PMC.
PMC approval of the release review is separate. Unless a
PMC specifies otherwise, the EMO regards a +1 from any PMC
member as approval. We employ about a day of "feet dragging"
to give PMC members an opportunity to challenge.
When Ivar cast a +1 for PMC in the ballot (and he
continues to do that), did that not cover PMC. Does EMO
need to be included in the ballot? Or, is there need to be
another mechanism for bring EMO into this process?
-- Ed
On 9/21/2020 11:02 AM, Scott Stark wrote:
> describes steps for a release review as
something that happens during
> the run up to ballot.
Right, it was brought up as an action item for
the EF to look into why they have diverged. It does
not apply to EE9 as every spec requires a release
review.
> The
interaction with the PMC and EMO have been described
as asynchronous
> by Wayne Beaton in the past. If those are
requirements for ballot
> completion, that needs to be made part of the
process.
Correct me if I'm wrong but as I understand it:
Ballot - an approval from the Jakarta EE WG for the
release/a process
defined by the spec committee
Release review - an approval from Eclipse Foundation
for the release/a
process defined by Eclipse Foundation
* both can run in parallel; EF/EMO will wait for
ballot to conclude and
approve the release if and only if ballot passes
* each checks different things (ie IP log, vendors'
expectations,...)
* one needs to get both approvals in order to
publish the release
Yes, that is how it is described in the https://www.eclipse.org/projects/efsp/,
but the checklist has no join point for this trio of
approvals: "A Release Review concludes
successfully with approval from the PMC and EMO,
and approval by a Super-majority of the
Specification Committee."
So really, the current 'on ballot completion, the
specification committee mentor...' checklist needs
to include waiting for approval from the EMO and PMC
before creating the project checklist item that
include the release of the APIs. Likewise, the
promotion of the TCK and merge of the specification
page should not be done by the mentor before this
approval.
I'll raise this on this week's spec committee
call and see what needs to be done about BV.